A step-by-step, beginner-friendly explanation of how a MOSFET transistor physically turns on and off, using the gate-source-drain model and everyday analogies rather than semiconductor physics equations.
The previous article introduced the transistor as "an electrically controlled switch." This article goes one level deeper and explains, in plain language with no physics equations, how that switching actually happens inside the silicon.
The Three Terminals: Gate, Source, and Drain
The most common transistor type used in modern chips is called a MOSFET (metal-oxide-semiconductor field-effect transistor). It has three terminals: the source, the drain, and the gate. A useful analogy is a water valve: the source and drain are like the pipe on either side of the valve, and the gate is the valve handle. Current wants to flow from source to drain, but whether it actually can flow is entirely controlled by the voltage applied to the gate — just as water can only flow through the valve when the handle is turned open.
Why Silicon Needs "Doping" to Work as a Switch
Pure silicon is a poor conductor on its own — it's neither a good conductor like copper nor a good insulator like glass, which is exactly why it's called a semiconductor. To make it useful as a controllable switch, manufacturers deliberately introduce tiny, precisely controlled amounts of other elements into the silicon crystal, a process called doping.
Adding an element like phosphorus (which has one extra outer electron compared to silicon) creates n-type silicon, with extra free electrons available to carry current. Adding an element like boron (which has one fewer outer electron) creates p-type silicon, leaving behind "holes" — missing electrons that behave, for circuit purposes, like mobile positive charges. A single transistor is built from specific, adjacent regions of n-type and p-type silicon, and it's the electrical behavior at the boundary between them that gives a transistor its switching ability.
What Happens When You Turn the Gate "On"
In an n-type MOSFET (the version most commonly used in digital logic, often paired with its p-type counterpart), the source and drain are both n-type silicon regions, separated by a thin strip of p-type silicon underneath the gate — a region with no free electrons readily available to carry current, which normally blocks any current from flowing between source and drain.
The gate sits just above this blocking region, separated from the silicon by an extremely thin insulating layer. When enough positive voltage is applied to the gate, it electrically attracts free electrons up into that thin p-type strip beneath it, temporarily forming a narrow conductive path — called a channel — connecting the source to the drain. With that channel formed, current can flow: the switch is "on." Remove the gate voltage, and the channel disappears; the blocking region returns, and current stops: the switch is "off." No mechanical motion happens anywhere in this process — it's purely electrons being electrically pulled into, or released from, that channel region, which is exactly why a transistor can switch billions of times per second with essentially no wear.
Putting It Together: Two Transistor Types, One Logic Gate
Modern digital chips almost universally use a design called CMOS (complementary metal-oxide-semiconductor), which pairs an n-type MOSFET with a p-type MOSFET (the electrical mirror image, switching on when its gate voltage is low instead of high) for every logic function. Wiring one of each type together so that exactly one of the pair is "on" at any given moment builds a basic logic gate — for example, a NOT gate, called an inverter in CMOS design, which outputs "off" when its input is "on" and vice versa. This complementary pairing is also why CMOS chips are so power-efficient: because only one transistor in the pair conducts at a time, almost no current flows straight through the gate except during the brief instant it's actually switching.
Why This Matters Even If You Never Design a Chip Yourself
Understanding this gate-source-drain switching mechanism is what makes the rest of chip design make sense. When you read that a chip has "billions of transistors," you now know exactly what that means: billions of these tiny gate-controlled channels, each one either letting current through or blocking it, combined into logic gates, and those logic gates combined into everything from a simple counter to a full AI processor. If you want to see this idea applied at the level of an actual chip design flow, the Build Your Own Silicon Chip simulator is a hands-on next step, and Semiconductor Device Physics Basics covers the same switching behavior with more technical depth for readers ready to go further.